Synthesis, antiproliferative activity and molecular docking of thiocolchicine urethanes

Article abstract of DOI:10.1016/j.bioorg.2018.09.004

A number of naturally occurring compounds such as paclitaxel, vinblastine, combretastatin, and colchicine exert their therapeutic effect by changing the dynamics of tubulin and its polymer form, microtubules. The identification of tubulin as a potential target for anticancer drugs has led to extensive research followed by clinical development of numerous compounds from several families. In this paper we report on the design, synthesis and in vitro evaluation of a group of thiocolchicine derivatives, modified at ring-B, labelled here compounds 4–14. These compounds have been obtained in a simple reaction of 7-deacetyl-10-thiocolchicine 3 with eleven different alcohols in the presence of triphosgene. These novel agents have been checked for anti-proliferative activity against four human cancer cell lines and their mode of action has been confirmed as colchicine binding site inhibition (CBSI) using molecular docking. Molecular simulations provided rational tubulin binding models for the tested compounds. On the basis of in vitro tests, derivatives 4–8 and 14 demonstrated the highest potency against MCF-7, LoVo and A549 tumor cell lines (IC₅₀ values = 0.009–0.014 μM). They were more potent and characterized by a higher selectivity index than several standard chemotherapeutics including cisplatin and doxorubicin as well as unmodified colchicine. Further, studies revealed that colchicine and its several derivatives arrested MCF-7 cells in mitosis, while its selected derivatives caused microtubule depolymerization.

Full text of DOI:10.1016/j.bioorg.2018.09.004

Accepted Manuscript
Synthesis, antiproliferative activity and molecular docking of thiocolchicine
urethanes
Urszula Majcher, Alicja Urbaniak, Ewa Maj, Mahshad Moshari, Magdalena
Delgado, Joanna Wietrzyk, Franz Bartl, Timothy C. Chambers, Jack A.
Tuszynski, Adam Huczyński
PII:
S0045-2068(18)30569-8
DOI:
https://doi.org/10.1016/j.bioorg.2018.09.004
YBIOO 2501
Reference:
Bioorganic Chemistry
To appear in:
Received Date:
Revised Date:
Accepted Date:
11 June 2018
5 September 2018
6 September 2018
Please cite this article as: U. Majcher, A. Urbaniak, E. Maj, M. Moshari, M. Delgado, J. Wietrzyk, F. Bartl, T.C.
Chambers, J.A. Tuszynski, A. Huczyński, Synthesis, antiproliferative activity and molecular docking of
thiocolchicine urethanes, Bioorganic Chemistry (2018), doi: https://doi.org/10.1016/j.bioorg.2018.09.004
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Bioorganic Chemistry- Regular Articles
Synthesis, antiproliferative activity and molecular docking of
thiocolchicine urethanes
Urszula Majchera, Alicja Urbaniakb, Ewa Majc, Mahshad Mosharid, Magdalena Delgadob,
Joanna Wietrzykc, Franz Bartle, Timothy C. Chambersb, Jack A. Tuszynskid, Adam
Huczyński*,a,
a
Faculty of Chemistry, Adam Mickiewicz University, Umultowska 89b, 61-614 Poznan,
Poland
b Department of Biochemistry and Molecular Biology, University of Arkansas for Medical
Sciences, Little Rock, AR 72205, United States of America
c
Hirszfeld Institute of Immunology and Experimental Therapy, Polish Academy of Sciences,
Rudolfa Weigla 12, 53-114 Wrocław, Poland
d Department of Oncology, University of Alberta, Edmonton, Alberta T6G 1Z2, Canada
e Institut für Biologie, AG Biophysikalische Chemie,– Humboldt Universität zu Berlin,
Invalidenstr, 42, 10099 Berlin, Germany
Abstract
A number of naturally occurring compounds such as paclitaxel, vinblastine, combretastatin,
and colchicine exert their therapeutic effect by changing the dynamics of tubulin and its
polymer form, microtubules. The identification of tubulin as a potential target for anticancer
drugs has led to extensive research followed by clinical development of numerous compounds
from several families. In this paper we report on the design, synthesis and in vitro evaluation
of a group of thiocolchicine derivatives, modified at ring-B, labelled here compounds 4-14.
These compounds have been obtained in a simple reaction of 7-deacetyl-10-thiocolchicine 3
with eleven different alcohols in the presence of triphosgene. These novel agents have been
checked for anti-proliferative activity against four human cancer cell lines and their mode of
action has been confirmed as colchicine binding site inhibition (CBSI) using molecular
docking. Molecular simulations provided rational tubulin binding models for the tested
compounds. On the basis of in vitro tests, derivatives 4-8 and 14 demonstrated the highest
potency against MCF-7, LoVo and A549 tumor cell lines (IC50 values = 0.009-0.014 μM).
They were more potent and characterized by a higher selectivity index than several standard
chemotherapeutics including cisplatin and doxorubicin as well as unmodified colchicine.
Further, studies revealed that colchicine and its several derivatives arrested MCF-7 cells in
mitosis, while its selected derivatives caused microtubule depolymerization.
1

* Corresponding author: adhucz@amu.edu.pl, (A. Huczyński)
Key words: Colchicine derivatives; Mitotic inhibitor; Anticancer agents; Colchicine binding
site inhibitors (CBSI); Tubulin.
1. Introduction
Colchicine (1), (see Scheme 1) - a major alkaloid isolated from Colchicum autumnale
and Gloriosa superba, has been in clinical use since 1810 for the treatment of acute gout and
is also used for familial Mediterranean fever and other diseases [1-2]. More recently, its
relatively high antiproliferative activity generated interest in the development of colchicine
and its derivatives as potential anti-cancer drugs [3-5]. The anticancer activity of 1 is related
to the formation of a colchicine–tubulin complex, which prevents microtubule polymerization
due to a conformational inflexibility making tubulin dimers incompetent for microtubule
assembly. As a consequence, the cells exposed to it tend to undergo mitotic arrest during the
cell cycle, followed by apoptosis [6-8]. Various families of microtubule-targeting agents have
been successfully applied in the treatment of several types of cancer [9-10]. However, the
relatively high toxicity of colchicine has prevented its use in cancer therapy thus far [5]. To
overcome colchicine’s general toxicity problem, numerous colchicine derivatives having high
potency and reduced toxicity have been synthesized and tested over the past few decades [8,
11-15]. From among a large number of synthetic colchicine analogues, thiocolchicine (2)
(Scheme 1) merits special attention. It has a C-10 methylthio group at the tropolone ring C
instead of a methoxy group and binds at the colchicine site on tubulin. Previous studies have
indicated that the replacement of oxygen by the sulfur atom leads to increased potency of the
compound [16-18]. Moreover from the chemical point of view 2 is also highly stable under
acidic hydrolytic conditions [16-18].
2

The aim of this work was to develop more selective thiocolchicine derivatives while
maintaining the potent antiproliferative activity of colchicine. The designed compounds (4-
14) shown in Figure 2 are carbamates of 7-deacetyl-10-thiocolchicine (3). We have chosen
carbamate derivatives (urethane) of 7-deacetyl-10-thiocolchicine (3) since it has been proved
that the carbamate group is a key structural motif in many approved drugs and prodrugs. In
fact, the use of carbamate derivatives has been consistently increasing in medicinal chemistry
and many derivatives have been specifically designed to cause drug–target interactions
through their carbamate moiety [19]. Additionally, some previously described carbamates of 1
have been also shown to be active against gouty arthritis [20].
Herein, we report the synthesis of a series of thiocolchicine analogues modified at
ring-B and obtained in the reaction of 7-deacetyl-10-thiocolchicine (3) with eleven different
alcohols in the presence of triphosgene. We also describe the results of in vitro
antiproliferative activity evaluation of the obtained derivatives against human cancer cell
lines, namely MCF-7, LoVo, LoVo/DX and A-549 as well as normal cells BALB/3T3. To
gain more knowledge about the molecular mechanism of action of the synthesized
compounds, we also present results of a molecular docking study, flow cytometric cell cycle
analyses, and investigation of the effects on microtubules organization using
immunofluorescence.
2. Results and discussion
2.1 Chemistry
Syntheses of the designed colchicine derivatives coupled to 3, with the structurally
different urethane moiety, were followed by investigation of their biological effects. The
synthetic route to the novel compounds encompassed two traditional synthetic modifications
and a new type of chemical modification, which is depicted in Scheme 1. The first two steps
3

of this synthesis are well described in literature. The first step was the preparation of
thiocolchicine from the starting material (colchicine) according to the reported procedure
[21]. Thiocolchicine (2) (see Scheme 1) has been readily available from colchicine by
treatment with sodium methanethiolate [21], and is a potent inhibitor of tubulin
polymerization resulting in the arrest of cell growth and cell division. Notably, thiocolchicine
binds to tubulin more rapidly than colchicine [22]. Hydrolysis of 2 with 20% HCl yielded 7-
deacetyl-10-thiocolchicine (3) [21].
In recent years, much attention has been focused on the synthesis of carbamates using
chloroformates as key intermediates. Carbamate synthesis through the chloroformate route
has been achieved using various reaction conditions, such as the use of organic and inorganic
bases. This method is not adequate for our designed thiocolchicine derivatives because the
respective chloroformate are not commercially available and some of them cannot be
synthesized [23]. Therefore, we searched for a cleaner and more efficient reaction. After
reviewing the available literature, triphosgene [bis(trichloromethyl) carbonate, BTC], a solid
and safe surrogate of phosgene [24], appeared as the reagent of choice for preparing
carbamates.
Subsequently, our one-pot synthesis of eleven carbamates 4-14 was achieved through
the reaction of amine 3 with a respective alcohol using triphosgene as the source of carbonyl
equivalent. Carbamates 4-14 (Scheme 1) were readily prepared from (3) and 1 equiv. of
triphosgene (3 fold molar excess of phosgene) in THF in the presence of triethylamine (10
equiv.). and then reacted with suitable primary or secondary alcohols (10 equiv.) to provide a
respective carbamate 4-14 with good yields in the range 41-90% (Scheme 1). To facilitate the
structural activity relationship analysis (SAR) we chose the primary and secondary alcohol
with different substituents such as: saturated alkyl chains (4-8), alkyl chains containing
additional OH group (9-11), alkyl chains containing oxygen atoms (10-13), alkyl chains
4

containing fluoride atoms (14). All obtained compound were easily isolated in pure form after
dry column vacuum chromatography on silica gel.
Scheme 1. Synthesis of colchicine derivatives (2-14). Reagents and conditions: (a) MeOH/H2O,
CH3SNa, RT; (b) 20% HCl, MeOH, reflux; (c) triphosgene, THF, Et3N, 0°C; primary or secondary
alcohols, RT.
1
13
The structures of all products 2-14 were determined using EI MS, FT-IR, H and C
NMR methods and are shown in Supplementary data and discussed below. The
13
1
characteristic signal of OCH3 group at C(10) position of (1) in the C and H NMR spectra
was observed at 56.1 ppm and as a singlet at 4.03 ppm, respectively. These signals vanish
completely (see spectra of compounds 2-14) after the reaction of (1) with sodium
methanethiolate proving the replacement of OCH3 group with SCH3 substitute at C(10)
position in the tropolone ring of 1. The signals of thiomethyl group in C(10) positions in the
1
13C NMR and H NMR spectra of compounds 2-14 are present at 15.1 ppm and
13
approximately 2.45 ppm, respectively. The characteristic C NMR signals assigned to the
carbon atoms in the acetamide group at C(7) position of unmodified 1 are observed at 170.3
ppm, and 22.7 ppm, respectively. These signals are no longer observed in the 13C NMR
5

spectrum of compound 3, which confirms hydrolysis of the acetyl group. The NH2 group of
compound 3 reacts easily with triphosgene and the primary or secondary alcohols giving
products 4-14 whose structures are confirmed by the presence of the 13C NMR signal
assigned to the carbon atom of the urethane group at 153.6-155.9 ppm and additionally by a
series of signals assigned to the aliphatic carbon atoms of the alcohol. In the 13C NMR spectra
of compounds 2-14 a signal assigned to C(7) carbon atom of B ring was observed at 56.0-
56.1 ppm, while in the spectrum of 1 it was observed at 52.8 ppm. The EI mass spectrometry
confirmed the structure of the obtained compounds by the presence of an m/z signal assigned
to the molecular ion of these compounds. The products of the chemo-selective one-pot
reaction 3 with triphosgene and alcohol were very well identified using also the FT-IR
spectroscopic method. The band assigned to the (C=O) vibrations (amide I band) of the
acetamide group in the spectrum of 2 at 1660 cm1, vanishes completely in the spectrum of 3,
indicating the absence of acetamide group with the formation of respective amine 3. The
bands assigned to the (C=O) vibrations of urethane group in the FT-IR spectra of 4-13 are
present in the range of 1710-1727 cm1. Only for compound 14 was this band shifted toward
1740 cm1 due to a strong inductive effect in the CH2CF3 moiety.
2.2 Inhibition of human cancer cell line growth
The synthesized 7-deacetyl-10-thiocolchicine derivatives 4-14 and starting compounds
1-3 were evaluated for their in vitro antiproliferative effect towards both normal and cancer
cells. Each compound was tested towards four human cancer cell lines, namely lung
adenocarcinoma (A549), breast adenocarcinoma (MCF-7), colon adenocarcinoma cell line
(LoVo) and its doxorubicin resistant subline (LoVo/DX). Antiproliferative effects were also
studied using a normal murine embryonic fibroblast cell line (BALB/3T3) in order to obtain
insight into their selectivity for cancer versus normal cells. IC50 values of the tested
6

compounds are presented in Table 1. To evaluate the activity against the cells with MDR
(multidrug resistance) phenotype, LoVo/DX cells were also tested and the indices of
resistance (RI) were calculated (see Table 1). The RI values indicate how many times more
resistant is the subline in comparison to its parental cell line.
Most of the obtained urethane derivatives of 7-deacetyl-10-thiocolchicine were more
potent, i.e. had lower IC50 values, than unmodified 1, intermediate compounds 2 and 3, as
well as doxorubicin and cisplatin, currently widely used as antitumor agents in cancer
chemotherapy. From the set of tested substances, the most active compounds against MCF-7,
LoVo, LoVo/DX and A549 tumor cell lines were 4-8 and 14 (IC50 = 0.009-0.010 μM).
Compound (11) showed the weakest activity against all cancer cell lines tested.
The calculated values of RI (Table 1) showed that the doxorubicin-resistant subline
(LoVo/DX) showed various levels of resistance to the parent compound and all of its
derivatives, which was consistent with them acting as substrates for MDR1-mediated drug
transport. These cells also exhibited resistance to doxorubicin, another MDR1 substrate, but
were actually slightly more sensitive to cisplatin than the parental cells (RI = 0.8), which was
consistent with cisplatin not acting as a substrate for MDR1-mediated transport [25].
Selectivity index (SI) is a major challenge in drug discovery, because it defines the
ability of a particular compound to preferentially kill tumor cells in relation to normal cells.
For this purpose, the obtained compounds were tested not only against tumor cells but also
against normal murine embryonic fibroblast cell line (BALB/3T3). The best values of the
selectivity index (see Table 2) for the tested compounds were obtained towards MCF-7 cells
(SI =1.89 -20.81). The best selectivity was observed for compound 7 which also showed high
antiproliferative activity against the four tested human cancer cell lines. The preliminary
structure-activity relationships (SARs) revealed that the type of substituent at C(7) position in
7

ring B and the presence of SCH3 group at C(10) position in ring C of colchicine (1) were of
cardinal importance to the compounds’ activity.
8

Table 1. Antiproliferative activity (IC50) [µM] of 1 and its derivatives 2-14 compared with that of standard anticancer drugs doxorubicin and
cisplatin. The calculated values of the resistance index (RI) for LoVo/DX versus LoVo cells are also shown.
Compound
MCF-7
LoVo
LoVo/DX
A-549
BALB/3T3
IC50 ±SD
IC50 ±SD
IC50 ±SD
RI
IC50 ±SD
IC50 ±SD
0.012 ±0.002
0.201 ±0.113
1.566 ±0.454
7.8
0.103 ±0.015
0.195 ±0.025
1
0.010 ±0.002
0.014 ±0.002
0.009 ±0.001
0.008 ±0.003
0.009 ±0.000
0.009 ±0.001
0.009 ±0.001
0.010 ±0.001
0.010 ±0.001
0.069 ±0.011
0.009 ±0.001
0.010 ±0.002
0.009 ±0.001
10.700 ±0.753
0.386 ±0.118
0.021 ±0.006
0.017 ±0.004
0.010 ±0.003
0.011 ±0.005
0.011 ±0.002
0.014 ±0.005
0.011 ±0.003
0.191 ±0.158
1.107 ±0.907
2.754 ±2.058
0.042 ±0.012
0.250 ±0.135
0.011 ±0.006
3.767 ±0.390
0.681 ±0.633
0.398 ±0.075 19.0
0.011 ±0.001
0.024 ±0.003
0.010 ±0.001
0.010 ±0.001
0.009 ±0.001
0.009 ±0.001
0.009 ±0.001
0.080 ±0.007
0.085 ±0.008
0.158 ±0.038
0.012 ±0.002
0.083 ±0.008
0.008 ±0.001
6.367 ±1.413
0.258 ±0.044
0.114 ±0.072
0.223 ±0.032
0.016 ±0.001
0.017 ±0.001
0.043 ±0.036
0.153 ±0.035
0.031 ±0.022
0.197 ±0.022
0.156 ±0.022
1.441 ±0.251
0.139 ±0.001
0.135 ±0.001
0.110 ±0.064
8.600 ±2.923
1.012 ±0.868
2
0.145 ±0.021
0.063 ±0.012
0.065 ±0.009
0.063 ±0.007
0.098 ±0.026
0.068 ±0.002
1.555 ±0.386
8.079 ±1.788
18.816 ±3.262
8.6
6.3
6.2
5.8
7.1
6.0
8.1
7.3
6.8
3
4
5
6
7
8
9
10
11
0.562 ±0.141 13.3
2.987 ±1.062 11.9
12
13
0.064 ±0.014
3.033 ±0.637
6.0
0.8
14
Cisplatin
Doxorubicin
8.187 ±1.409 12.0
The IC50 value is defined as the concentration at which 50% growth inhibition is observed. The following cell lines were used: human lung adenocarcinoma (A549), human
breast adenocarcinoma (MCF-7), human colon adenocarcinoma (LoVo) and doxorubicin resistant subline (LoVo/DX), normal murine embryonic fibroblast (BALB/3T3).
The RI (Resistance Index) indicates fold-resistance relative to parental cell line. The RI was calculated for each compound using the formula: RI = IC50 for LoVoDX / IC= for
LoVo cell line. RI < 1, MDR cells are more sensitive; RI = 1–2, no or very low level of resistance; RI = 2–10, moderate resistance; RI >10, strong drug-resistance.
9

Table 2. Selectivity indexes
Compound
Calculated selectivity index SI
MCF-7
LoVo/Dx
0.12
0.29
1.54
0.26
0.26
0.68
1.56
0.45
0.13
0.02
0.08
0.25
0.05
1.72
2.84
0.12
LoVo
0.98
5.45
13.21
1.63
1.63
3.95
11.11
2.74
1.03
0.14
0.52
3.29
0.54
10.38
2.28
1.49
A-549
1.90
10.08
9.25
1.63
1.78
4.94
16.28
3.37
2.45
1.84
9.09
11.33
1.63
13.74
1.35
3.93
1
16.71
11.85
15.71
1.89
2
3
4
5
2.07
6
4.94
7
16.28
3.37
8
9
19.28
15.80
20.81
15.82
13.99
11.70
0.80
10
11
12
13
14
Cisplatin
Doxorubicin
2.62
The SI (Selectivity Index) was calculated for each compound using the formula: SI= IC50 for a normal cell line
BALB3T3/IC50 for a respective cancerous cell line. A beneficial SI > 1.0 indicates a drug with efficacy against
tumor cells greater than the toxicity against normal cells. The bolded numbers indicate compounds characterized
by beneficial SI index and concomitant low IC50 towards respective cancer cell line.
The compounds with saturated alkyl chains (4-8), and alkyl chains containing fluoride
atoms (14) displayed more potent antiproliferative activities than those with alkyl chains
containing an additional OH group (9-11) or alkyl chains containing oxygen atoms (9-13).
Compounds 7 and 14 with short and branched substituent (isopropyl and trifluoroethyl) show
the highest ability to inhibit proliferation of the cancer cell lines tested. Simultaneously, their
activities are higher than those of the chemically unmodified colchicine and thiocolchicine as
well as doxorubicin and cisplatin, commonly used as antitumor agents in cancer
chemotherapy. To summarize these studies, all obtained derivatives have been found active,
to varying degrees, in proliferation inhibition in a specified concentration range. Their
activities depend on the tested cancer cell line. Compounds 7 and 14 with the best anti-
proliferative properties and good selectivity towards cancer cells are promising anticancer
candidates.
10

2.3. Molecular docking: in silico determination of the molecular mode of action
In order to determine the molecular mode of action of the new thiocolchicine
derivatives we performed computational modeling aimed at the ligand-protein interaction. As
the target protein we used the most common tubulin isoform, βІ tubulin, which is one of the
subunits of microtubules. The structures of the new thiocolchicine derivatives described
above were docked into the βІ tubulin colchicine binding site and ranked according to their
binding affinity. More specifically, a combination of different computational methods was
used to explore ligand-tubulin interactions. The ligand structures were first minimized and
then fully optimized on the basis of the RHF/cc-pVDZ level of theory in the GAMESS-US
software, version 2010-10-01 (http://www.msg.ameslab.gov/gamess/). Since there is no
crystal structure available of human βІ tubulin (TBB5_HUMAN), we first obtained its amino
acid sequence from UniProt (http://www.uniprot.org/; ID: Q13509). Next, we used the Protein
Data Bank (https://www.rcsb.org/) to obtain the tubulin structure 1SA0.pdb as a template to
construct a homology model for βІ tubulin using the Molecular Operating Environment
software (MOE, 2015 https://www.chemcomp.com/). We then docked our thiocolchicine
library of derivatives to the target protein using the Autodock 4.2 program
(http://autodock.scripps.edu/) and applied flexible ligand and rigid receptor conditions.
AutoDock 4.2 software is designed to predict how drug candidates bind to a receptor of a
known 3D structure, and consists of two main programs: autodock that performs the docking
of the ligand to a set of grids describing the target protein and autogrid that pre-calculates
these grids.
Binding energies interactions (BE) between the 14 different thiocolchicine derivatives
and βІ tubulin and the estimated Moriguchi octanol-water partition coefficient, MlogP for
each colchicine derivative are shown in Table 3. In order to arrive at a structure-function
relationship for the series of compounds obtained, we also calculated the molecular weight of
11

each compound and the length of the corresponding aliphatic chain. Most importantly, the
binding energy for each compound was determined and we presented graphically the ligand-
protein interactions showing the amino acids of the protein and the atomic groups of the
ligand that engage in physical and chemical interactions, whose graphical representation is
explained in the legend of Table 3 where all the computational results are summarized.
Table 3. Computational predictions of the interactions between thiocolchicine derivatives and βІ
tubulin. Binding energy values in kcal/mol, graphical representation of the ligand-protein interactions,
molecular weight, partition coefficient and the length of the aliphatic chains involved are tabulated.
Binding
Energy
(BE)
Length
of
aliphatic
chain
Comp.
Graphical representations of the interactions
MW
MlogP
(kcal/mol)
399.45
-8.09
1.37
5.841
1
-8.13
421.56
1.56
5.841
2
12

-7.56
379.52
1.57
2.035
3
-8.59
437.56
1.61
4.654
4
-8.57
451.58
1.83
4.945
5
13

-8.55
465.61
2.04
6.790
6
-8.67
465.61
2.05
4.898
7
-8.19
479.64
2.25
8.342
8
14

-7.91
467.58
1.08
6.479
9
-7.36
511.64
0.76
8.645
10
-6.51
555.69
0.44
12.433
11
15

-8.01
495.64
1.50
9.277
12
-7.53
519.61
0.97
10.425
13
-7.67
505.56
2.14
4.588
14
16

It is clear from the results summarized in Table 3 that all these compounds show high
affinity to the colchicine binding site on  tubulin. It should be mentioned that the values of
binding energies are not absolute but relative, to establish a rank order of the compounds. The
strongest binding compound is predicted to be 7, followed closely by 4, 5, 6 and 8. On the
other end of the spectrum, compound 11 appears to be the weakest binder to the target. The
most interactive residues seem to be Lys 352, Lys 254, Asn 258, Leu 255, Leu 248, Met 259
and Cys 241 which appear consistently involved in the binding of the thiocolchicine
derivatives. To a less degree, Asp 251 and Thr 353 also appear to play a role in several cases.
In Fig. 1 we show the mode of action of the compounds studied in this paper for the
compounds divided into two groups: (a) the ones that bind to tubulin with binding energies of
-8 kcal/mol and higher and (b) the ones that bind to tubulin with binding energies lower than
-8 kcal/mol.
(a)
(b)
Fig. 1. Illustration in 3D of the interaction modes of the tubulin βI structure with: (a) 8 thiocolchicine
derivatives with binding energies of -8 kcal/mol and higher; (b) with 6 thiocolchicine derivatives with
binding energies lower than -8 kcal/mol.
To better understand the structure-activity relationship for this series of compounds,
we performed linear regression analysis with respect to the binding energy (Table 3) and
17

MlogP (Table 3) versus the experimentally determined values of IC50 (Table 1). The results
of this analysis are shown in Figures S14-S18 (Supplementary material).
The 3D plots indicate that there is a reasonable degree of correlation between the
experimental activity data in terms of IC50 values for the cell lines studied and the
computational prediction data expressed as the binding energy and partition coefficient. To
quantify these relationships, in Table 4 we summarize the linear regression coefficients for
these relationships for each cell line separately. The values of the regression coefficient (R2)
range from 0.43 for MCF-7 (which is a very aggressive breast cancer cell line) and 0.88 for
LoVo/DX.
Table 4. Linear regressions two independent (binding energy and MlogP) and log IC50 [μM].
MCF-7
0.43
LoVo/DX
0.88
LoVo
0.86
A-549
0.74
BALB/3T3
0.55
Linear
regression (R2)
2.4. Colchicine and its analogs induce G2/M arrest in MCF-7 cells
In order to examine the effect of colchicine 1 and the most potent starting compounds 2-3 and
derivatives 4-6 and 14 on the cell cycle progression, flow cytometric cell cycle analysis after
propidium iodide staining was performed. MCF-7 cells were treated with 0.1% DMSO
(vehicle) or compounds 1-6 and 14 at concentrations equal to 10 x IC50 values determined via
cell viability assay (Table 1) for 24, 48 and 72 h. The full set of representative cytograms is
included in Supplementary material (Fig. S19). A graphical representation of summarized
percentage of cells in G2/M phase of the cell cycle (n=3) is shown in Fig. 2. For all of the
compounds studied, statistically significant G2/M phase arrest was observed as early as 24 h
after drug addition, with the level of arrested cells slightly decreasing after further incubation
for 48 and 72 h Fig. 2.
Supplementary
material
18

Fig. 2. Colchicine (1), starting compounds (2-3) and derivatives (4-6, 14) induced G2/M arrest
in MCF-7 cells. MCF-7 cells were treated with the indicated compounds for 24, 48 or 72 h
and subjected to propidium iodide staining and flow cytometry. Percent of cells with 4N DNA
(G2/M phases) was determined (mean ± SD, n = 3). (*** p < 001, ** p < 0.01, * p < 0.05 vs.
control). See Supplementary material for a full set of representative data.
2.5. Effect of colchicine and compound 5 on microtubule depolymerization
Immunofluorescence staining for β-tubulin in MCF-7 cells was performed to investigate the
effect of 1) and compound 5 (selected based on having the lowest IC50 value for MCF-7 cells,
Table 1) on microtubule integrity. HeLa cells treated with vincristine (VCR) served as a
positive control for microtubule depolymerization, as demonstrated previously [26-27].
Untreated MCF-7 and HeLa cells exhibited distinct microtubule structures (Fig. 3 A and B,
first row) and as expected after treatment with VCR both cell types exhibited diffuse β-tubulin
staining with no detectable microtubules, which was consistent with depolymerization (Fig. 3
A and B, second row). Parent compound 1 and compound 5 also clearly induced microtubule
depolymerization, similar to the effects of VCR, in both HeLa and MCF-7 cells (Fig. 3 A and
B, third row and fourth row).
19

A.
DAPI
β-tubulin
Overlay
vehicle
VCR
1
5
20

B.
DAPI
β-tubulin
Overlay
vehicle
VCR
1
5
21

Fig. 3 Effects of 1 and 5 on microtubule structures in HeLa (A) and MCF-7 (B) cells. HeLa and MCF-
7 cells were incubated for 12 h without (vehicle, 0.1% DMSO) or with 1 (120 nM), 5 (80 nM)
(concentrations corresponding with 10 x IC50 values) or 100 nM VCR (positive control). Bar equals 5
µm (A) or 10 µm (B).
3. Conclusion
In conclusion, we designed and synthesized a new series of colchicine 1 derivatives
based on modification of previously reported tubulin polymerization inhibitors such as
thiocolchicine 2. Compounds 4-14 were prepared from colchicine by an efficient three-step
synthetic procedure including synthesis of 2 and its hydrolysis to 7-deacetyl-10-thiocolchicine
3. Eleven target carbamates of 7-deacetyl-10-thiocolchicine (4-14) were obtained using a
simple “one-pot reaction” of 3 with eleven different alcohols in the presence of triphosgene.
When tested for antiproliferative activity, most of the derivatives were more potent than
colchicine. An exception was compound 11, which had a higher IC50 than colchicine against
all tested cell lines. In terms of selectivity for tumor versus normal cells, many of the
derivatives were superior to colchicine. Compounds 7 and 14, for example, were highly
selective for three of the four tested cancer cell lines when compared to colchicine (Table 2)
and were also more potent than colchicine against all cancer cell lines (Table 1). Compounds
7 and 14 in particular, therefore represent promising leads for further development. The
results also reveal that the nature of the aliphatic chain at the urethane group and the nature of
the heteroatom (F better than O) have a marked influence on the antiproliferative activity of
the compounds. Generally, we observed also that the introduction of a functionalized alkyl
chain in the urethane group, for example by an end-standing hydroxyl group i.e. 9, 10, 11,
decreased the anti-proliferative activity and also that long alkyl chain including oxygen atoms
12, 13 did not improve this activity. Molecular docking was performed and revealed that all
of the obtained compounds successfully dock in the colchicine binding site of tubulin.
Therefore, these results suggest that carbamates of 7-deacetyl-10-thiocolchicine 4-14 are
22

potential inhibitors of tubulin polymerization. This conclusion was supported by studies on
select derivatives whose results showed that they caused mitotic arrest and promoted tubulin
depolymerization, as displayed by colchicines [5-10, 26]. Finally, these compounds are
amenable to further structural modifications and could serve as useful pharmacophore
templates for the generation of molecules as potential anticancer agents.
4. Experimental
4.1. General
All precursors for the synthesis of colchicine derivatives and solvents were obtained from
Aldrich or Fluka and were used as received without further purification. CDCl3 spectral–grade
solvents were stored over 3 Å molecular sieves for several days. All the solvents used in flash
chromatography were of HPLC grade (CHROMASOLV from Sigma–Aldrich) and were used
as received. Reaction mixtures were stirred using Teflon-coated magnetic stir bars. Reactions
were monitored by thin layer chromatography (TLC) using aluminum-backed plates (Merck
1
60F254). TLC plates were visualized in UV-light (254 nm). The H, 13C spectra were
recorded on a Varian VNMR-S 400 MHz spectrometer using TMS as the internal standard in
1
both cases. No window function or zero filling was used. H NMR measurements of 1-14
(0.07 mol dm-3) in CDCl3 were carried out at the operating frequency 402.64 MHz. The error
of the chemical shift value was 0.01 ppm. The 13C NMR spectra were recorded at the
operating frequency 101.25 MHz. The error of chemical shift value was 0.1 ppm. All spectra
were locked to deuterium resonance of CDCl3. The 1H and 13C NMR spectra are shown in the
Supplementary Materials. Then, the FT-IR spectra were recorded in the mid infrared region
in KBr pallets (2.0/200.0 mg) at 300 K on an IFS 113v FT-IR spectrophotometer (Bruker,
Karlsruhe) equipped with a DTGS detector; resolution 2 cm–1, NSS = 64. The Happ-Genzel
apodization function was used. Low-resolution EI-mass spectra were recorded using AMD-
23

Intectra GmbH (Harpstedt) D-27243 model 402 double-focusing sector mass spectrometer
(ionizing voltage 70 eV; accelerating voltage 8 kV; resolution 10,000; 10% valley definition).
The elemental analyses of 2–6 were carried out on Vario ELIII (Elementar, Germany).
4.2. Synthesis
General procedure for the synthesis of 7-deacetyl-10-thiocolchicine derivatives (4-14)
The synthetic routes of target compounds 4-14 are depicted in Scheme 1. A solution
7-deacetyl-10-thiocolchicine (3) (200 mg, 0.5355 mmol) in tetrahydrofuran (15 ml) and 0.7
cm3 triethyalmine were stirred at 0°C for 15 minutes. Triphosgene (158.9 mg, 0.5355 mmol)
dissolved in tetrahydrofuran (2 ml) was added dropwise to the cooled solution and stirred at
0°C for 30 minutes. The solution was then brought to room temperature and from 0.2 ml to 1
ml (5.355 mmol) was added of the suitable primary or secondary alcohols (for example, 0.22
ml methanol). The course of the reaction was followed by silica gel TLC. The mixture was
stirred at room temperature for 2 days and then filtered to remove the precipitated
triethylamine hydrochloride, and the remaining solution was evaporated under reduced
pressure. The resulting mixture was dissolved in CH2Cl2 and extracted with 2N HCl (aq), and
then solvated with water. The organic layers were combined and evaporated to dryness under
reduced pressure and purified using dry-column flash chromatography (Silica gel 60 from
Fluka; chloroform/acetone 10:2) to give respective urethanes derivatives of 7-deacetyl-10-
thiocolchicine 4-14.
Compound 4
1
Yield 203 mg, 88%; H NMR (403 MHz, CDCl3) δ 7.34 (s, 1H), 7.28 (d, J = 0.4 Hz, 1H),
7.26 (s, 1H), 7.05 (d, J = 10.4 Hz, 1H), 6.55 (s, 1H), 5.30 (s, 1H), 4.43 (dt, J = 13.3, 6.8 Hz,
1H), 3.94 (s, 3H), 3.91 (s, 3H), 3.64 (s, 3H), 3.58 (s, 3H), 2.54 (m, 1H), 2.44 (s, 3H), 2.28 (m,
1H), 1.84 – 1.74 (m, 1H);13C NMR (101 MHz, CDCl3) δ 182.4, 158.3, 155.9, 153.5, 151.1,
150.5, 141.6, 137.7, 134.5, 134.3, 128.6, 126.2, 125.6, 107.3, 61.4, 61.3, 56.0, 53.7, 52.2,
37.0, 29.9, 15.1; FT-IR (KBr pellet, νmax, cm−1) 3231, 2942, 1727, 1608, 1538, 1486, 1406,
24

1349, 1321, 1255., 1195, 1137, 1099, 1068, 1023, Anal. calcd. (%) for C22H25NO6S: C 61.24;
H 5.84; N 3.25; S 7.43, Found: C 61.12; H 5.9;3 N 3.21; S 7.29, MS (EI) m/z 431.3 (M)
Compound 5
1
Yield 200 mg, 84%; H NMR (403 MHz, CDCl3) δ 7.30 (s, 1H), 7.28 (d, J = 7.3 Hz, 1H),
7.25 (s, 1H), 7.04 (d, J = 10.5 Hz, 1H), 6.53 (s, 1H), 5.18 (d, J = 7.5 Hz, 1H), 4.42 (dt, J =
13.5, 6.8 Hz, 1H), 4.00 (qd, J = 7.3, 4.2 Hz, 2H), 3.93 (s, 3H), 3.90 (s, 3H), 3.63 (s, 3H), 2.56
– 2.48 (m, 1H), 2.43 (s, 3H), 2.33 – 2.21 (m, 1H), 1.74 (m, 1H), 1.19 (t, J = 7.1 Hz, 3H); 13C
NMR (101 MHz, CDCl3) δ 182.4, 158.2, 155.4, 153.5, 151.1, 150.6, 141.5, 137.7, 134.4,
134.3, 128.7, 126.2, 125.6, 107.3, 61.4, 61.3, 61.0, 56.0, 53.5, 37.1, 29.9, 15.1, 14.4; FT-IR
(KBr pellet, νmax, cm−1) 3225, 2937, 1718, 1608, 1537, 1486, 1406, 1374, 1348, 1321, 1284,
1256, 1139, 1097, 1065, 1025; Anal. calcd. (%) for C23H27NO6S: C 62.00; H 6.11; N 3.14; S
7.20, Found: C 62.12; H 6.03; N 3.06; S 7.03, MS (EI) m/z 445.2 (M)
Compound 6
1
Yield 177 mg, 72%; H NMR (403 MHz, CDCl3) δ 7.35 (s, 1H), 7.28 (d, J = 2.5 Hz, 1H),
7.26 (s, 1H), 7.05 (d, J = 10.5 Hz, 1H), 6.55 (s, 1H), 5.49 (s, 1H), 5.30 (s, 1H), 4.43 (dt, J =
13.3, 6.9 Hz, 1H), 3.94 (s, 3H), 3.91 (s, 3H), 3.64 (s, 3H), 2.54 (dd, J = 13.4, 6.0 Hz, 1H),
2.44 (s, 3H), 2.34 – 2.22 (m, 2H), 1.84 – 1.75 (m, 1H), 1.55 (dd, J = 14.2, 7.1 Hz, 2H), 0.87
13
(t, J = 7.4 Hz, 3H); C NMR (101 MHz, CDCl3) δ 182.4, 158.2, 155.5, 153.5, 151.1, 150.5,
141.6, 137.7, 134.4, 134.3, 128.7, 126.2, 125.6, 107.3, 66.7, 61.4, 61.3, 56.0, 53.5, 37.1, 29.9,
22.1, 15.1, 10.2; FT-IR (KBr pellet, νmax, cm−1) 3210, 2936, 1720, 1608, 1539, 1486, 1406,
1348, 1320, 1284, 1255, 1199, 1139, 1097, 1072, 1027; Anal. calcd. (%) for C24H29NO6S: C
62.73%; H 6.36%; N 3.05%; S 6.98%, Found C 62.66%; H 6.52%; N 3.01%; S 6.75%, MS
(EI) 459.2 (M).
Compound 7
Yield 86 mg, 35%; 1H NMR (403 MHz, CDCl3) δ 7.32 (s, 1H), 7.28 (d, J = 3.0 Hz, 1H), 7.26
(s, 1H), 7.05 (d, J = 10.5 Hz, 1H), 6.55 (s, 1H), 5.24 (s, 1H), 4.82 – 4.70 (m, 1H), 4.43 (dt, J =
11.8, 6.8 Hz, 1H), 3.94 (s, 3H), 3.91 (s, 3H), 3.64 (s, 3H), 2.65 – 2.48 (m, 1H), 2.44 (s, 3H),
2.33 – 2.23 (m, J = 19.4, 12.6, 6.5 Hz, 1H), 2.07 – 1.98 (m, 1H), 1.86 – 1.70 (m, 1H), 1.21 –
1.08 (m, 6H); 13C NMR (101 MHz, CDCl3) δ 182.4, 158.2, 154.9, 153.5, 151.1, 150.5, 141.6,
137.6, 134.4, 134.3, 128.7, 126.2, 125.6, 107.2, 68.5, 61.4, 61.3, 56.0, 53.4, 37.3, 29.9, 22.1,
22.0, 15.1; FT-IR (KBr pellet, νmax, cm−1) 3243, 2935, 1711, 1608, 1525, 1486, 1439, 1407,
25

1379, 1346, 1317, 1256, 1139, 1119, 1095, 1027; Anal. calcd. (%) for C24H29NO6S: C 62.73;
H 6.36; N 3.05; S 6.98, Found C 62.67; H 6.44; N 3.15; S 6.78, MS (EI) m/z 459.3 (M).
Compound 8
1
Yield 169 mg, 67%; H NMR (403 MHz, CDCl3) δ 7.38 (s, 1H), 7.29 (d, J = 1.1 Hz, 1H),
7.27 (s, 1H), 7.06 (d, J = 10.4 Hz, 1H), 6.55 (s, 1H), 5.68 (s, 1H), 4.43 (dt, J = 25.3, 9.5 Hz,
1H), 3.94 (s, 3H), 3.91 (s, 3H), 3.64 (s, 3H), 2.53 (d, J = 6.1 Hz, 1H), 2.44 (s, 3H), 2.30 (ddd,
J = 19.0, 12.4, 6.2 Hz, 1H), 2.05 – 1.75 (m, 2H), 1.82 (dd, J = 17.7, 11.4 Hz, 1H), 1.53 – 1.44
(m, 2H), 1.35 – 1.24 (m, 2H), 0.87 (t, J = 7.3 Hz, 3H); 13C NMR (101 MHz, CDCl3) δ 182.4,
158.2, 155.6, 153.5, 151.0, 150.7, 141.5, 137.7, 134.4, 134.3, 128.7, 126.2, 125.5, 107.2, 64.9,
61.4, 61.3, 56.0, 53.5, 37.1, 30.7, 29.8, 18.9, 15.0, 13.6; FT-IR (KBr pellet, νmax, cm−1) 3308,
2934, 1715, 1607, 1546, 1486, 1404, 1349, 1322, 1247, 1137, 1095, 1021 Anal. calcd. (%) for
C25H31NO6S: C 63.40; H 6.60; N 2.96; S 6.77, Found C 63.29; H 6.68; N 2.98; S 6.65, MS
(EI) m/z 473.2 (M).
Compound 9
1
Yield 222 mg, 90%; H NMR (403 MHz, CDCl3) δ 7.46 (s, 1H), 7.31 (d, 1H), 7.28 (s, 1H),
6.65 (d, J = 7.3 Hz, 1H), 6.56 (s, 1H), 4.40 (dt, 1H), 4.13 – 4.01 (m, 2H), 3.94 (s, 3H), 3.91 (s,
3H), 3.63 (s, 3H), 3.36 (s, 1H), 2.61 – 2.49 (m, 1H), 2.44 (s, 3H), 2.35 – 2.24 (m, J = 18.9, 6.1
13
Hz, 1H), 1.96 – 1.86 (m, 1H); C NMR (101 MHz, CDCl3) δ 182.5, 158.2, 155.9, 153.6,
151.3, 151.0, 141.5, 138.2, 134.7, 134.4, 128.7, 126.7, 125.5, 107.3, 66.8, 61.4, 61.3, 61.0,
56.0, 53.8, 36.8, 29.9, 15.1; FT-IR (KBr pellet, νmax, cm−1) 3314, 2936, 1715, 1606, 1540,
1486, 1404, 1349, 1322, 1238, 1195, 1137, 1095, 1021; Anal. calcd. (%) for C23H27NO7S: C
59.85; H 5.90; N 3.03; S 6.95, Found C 59.77; H 5.99; N 2.98; S 6.79, MS (EI) m/z 461.2.
Compound 10
1
Yield 119 mg, 44%; H NMR (403 MHz, CDCl3) δ 7.36 (s, 1H), 7.29 (d, J = 10.5 Hz, 1H),
7.27 (s, 1H), 7.07 (d, J = 10.5 Hz, 1H), 6.55 (s, 1H), 5.85 (s, 1H), 4.40 (dt, 1H), 4.30 – 4.15
(m, 2H), 4.16 – 4.02 (m, 2H), 3.94 (s, 3H), 3.91 (s, 3H), 3.74 (d, J = 4.4 Hz, 2H), 3.63 (s, 3H),
3.60 – 3.56 (m, 2H), 3.57 (s, 1H), 2.57 – 2.51 (m, 1H), 2.44 (s, 3H), 2.33 – 2.24 (m, 1H), 1.85
13
– 1.76 (m, 1H); C NMR (101 MHz, CDCl3) δ 182.4, 158.4, 155.3, 153.6, 151.1, 150.7,
141.6, 137.8, 134.7, 134.3, 128.5, 126.4, 125.5, 107.3, 72.4, 69.1, 64.1, 61.5, 61.4, 61.3, 56.1,
53.7, 36.9, 29.8, 15.1; FT-IR (KBr pellet, νmax, cm−1) 3312, 2934, 1716, 1606, 1543, 1486,
26

1459, 1404, 1349, 1322, 1252, 1137, 1095, 1021; Anal. calcd. (%) for C25H31NO8S: C 59.39;
H 6.18; N 2.77; S 6.34, Found C 59.44; H 6.07; N 2.67; S 6.15 MS (EI) m/z 505.3 (M).
Compound 11
1
Yield 235 mg, 78%; H NMR (403 MHz, CDCl3) δ 7.34 (s, 1H), 7.27 (d, J = 9.8 Hz, 1H),
7.26 (s, 1H), 7.06 (d, J = 10.6 Hz, 1H), 6.57 (d, J = 14.6 Hz, 1H), 6.12 (s, 1H), 4.40 (dt, J =
13.3, 6.8 Hz, 1H), 4.16 (dt, J = 11.8, 4.2 Hz, 2H), 4.20 – 4.00 (m, 2H), 3.94 (s, 3H), 3.92 (d, J
= 4.6 Hz, 4H), 3.91 (s, 3H), 3.81 – 3.73 (m, 2H), 3.70 – 3.67 (m, 2H), 3.65 – 3.65 (m, 1H),
3.63 (s, 3H), 2.57 – 2.49 (m, 1H), 2.44 (s, 3H), 2.32 – 2.21 (m, 1H), 1.84 – 1.77 (m, J = 11.6,
13
5.2 Hz, 1H); C NMR (101 MHz, CDCl3) δ 182.4, 158.2, 155.3, 153.5, 151.0, 150.6, 141.5,
137.7, 134.5, 134.3, 128.6, 126.3, 125.5, 107.2, 72.3, 70.4, 70.0, 69.2, 63.9, 61.5, 61.4, 61.3,
56.0, 53.6, 36.9, 29.8, 15.1; FT-IR (KBr pellet, νmax, cm−1) 3312, 2934, 1716, 1606, 1543,
1486, 1459, 1404, 1349, 1322, 1252, 1137, 1095, 1021; Anal. calcd. (%) for C27H35NO9S: C
59.00; H 6.42; N 2.55; S 5.83, Found C 59.18; H 6.36; N 2.64; S 5.65, MS (EI) m/z 549.0
(M).
Compound 12
1
Yield 111 mg, 41%; H NMR (403 MHz, CDCl3) δ 7.30 (s, 1H), 7.27 (d, J = 2.6 Hz, 1H),
7.25 (s, 1H), 7.04 (d, J = 10.7 Hz, 1H), 6.54 (s, 1H), 5.37 (s, 1H), 4.41 (dt, J = 11.7, 6.7 Hz,
1H), 4.18 – 4.07 (m, 2H), 3.94 (s, J = 2.0 Hz, 3H), 3.92 (s, 1H), 3.91 (s, 3H), 3.63 (s, 3H),
3.56 (ddd, J = 8.9, 5.4, 3.3 Hz, 2H), 3.53 – 3.48 (m, 2H), 2.60 – 2.48 (m, J = 13.6, 5.9 Hz,
1H), 2.43 (s, J = 4.1 Hz, 3H), 2.31 – 2.23 (m, 1H), 1.80 – 1.73 (m, J = 12.1, 6.5 Hz, 1H), 1.21
13
(t, J = 7.0 Hz, 2H); C NMR (101 MHz, CDCl3) δ 182.4, 158.3, 155.1, 153.5, 151.1, 150.2,
141.6, 137.5, 134.5, 134.2, 128.7, 126.2, 125.6, 107.3, 68.5, 66.6, 64.4, 61.4, 61.4, 56.1, 53.6,
37.2, 29.8, 15.1, 15.0; FT-IR (KBr pellet, νmax, cm−1) 3310, 2932, 1718, 1607, 1548, 1486,
1459, 1404, 1349, 1322, 1285, 1248, 1195, 1137,1095, 1068, 1022; Anal. calcd. (%) for
C25H31NO7S: C 61.33; H 6.38; N 2.86; S 6.55, Found C 61.27; H 6.48; N 2.76; S 6.51, MS
(EI) m/z 489.2 (M).
Compound 13
1
Yield 145 mg, 52%; H NMR (403 MHz, CDCl3) δ 7.29 (s, 1H), 7.27 (d, J = 2.6 Hz, 1H),
7.25 (s, 1H), 7.04 (d, J = 10.7 Hz, 1H), 6.54 (s, 1H), 5.33 (d, J = 7.3 Hz, 1H), 4.42 (dt, J =
11.8, 6.8 Hz, 1H), 4.18 (dd, J = 6.8, 5.1 Hz, 2H), 4.15 – 4.08 (m, 2H), 3.94 (s, 3H), 3.91 (s,
3H), 3.63 (s, 3H), 3.62 – 3.59 (m, 2H), 3.55 (td, J = 4.0, 1.9 Hz, 2H), 3.38 (s, 3H), 2.58 – 2.50
27

3
(m, 1H), 2.43 (s, 3H), 2.32 – 2.24 (m, 1H), 1.79 – 1.72 (m, 1H); C NMR (101 MHz, CDCl3)
δ 182.4, 158.3, 155.1, 153.5, 151.1, 150.1, 141.6, 137.5, 134.5, 134.2, 128.7, 126.1, 125.6,
107.3, 71.8, 70.4, 69.4, 64.2, 61.4, 61.4, 59.0, 56.1, 53.6, 37.2, 29.7, 15.1; FT-IR (KBr pellet,
νmax, cm−1) 3303, 2920, 1718, 1608, 1551, 1485, 1404, 1349, 1322, 1247,1195, 1137, 1095,
1021; Anal. calcd. (%) for C26H33NO8S: C 60.10; H 6.40; N 2.70; S 6.17, Found C 60.28; H
6.58; N 2.61; S 6.00, MS (EI) m/z 519.2 (M).
Compound 14
1
Yield 115 mg, 43%; H NMR (403 MHz, CDCl3) δ 7.47 (s, 1H), 7.30 (s, J = 10.4 Hz, 1H),
7.28 (s, 1H), 7.10 (d, J = 10.6 Hz, 1H), 6.62 (d, J = 4.9 Hz, 1H), 6.57 (s, 1H), 4.47 (dt, J = 4.0
Hz, 2H), 3.95 (s, 3H), 3.92 (s, 3H), 3.62 (s, 3H), 3.59 (s, 1H), 2.55 (s, 1H), 2.45 (s, 3H), 2.05
13
– 2.01 (m, 1H), 1.89 (s, 1H); C NMR (101 MHz, CDCl3) δ 182.4, 158.6, 153.6, 153.6,
151.0, 150.3, 141.6, 137.7, 134.8, 134.3, 128.8, 126.5, 125.4, 107.3, 61.3, 61.2, 60.9, 60.6,
56.0, 54.1, 36.9, 29.8, 15.1; FT-IR (KBr pellet, νmax, cm−1) 3314.3, 2938, 1740, 1608, 1546,
1486, 1350, 1322, 1237, 1163, 1097, 1022; Anal. calcd. (%) for C23H24F3NO6S: C 55.30; H
4.84; N 2.80; S 6.42, Found C 55.15; H 4.98; N 2.80; S 6.14, MS (EI) m/z 499.1 (M).
1
The H, 13C NMR, FT-IR and EI MS spectra all tested compounds are included in the
Supplementary material (Figs S1- S13).
4.3. Antiproliferative activity of colchicine and its derivatives
A549, MCF-7 cell lines were obtained from the European Collection of Authenticated Cell
Cultures (ECACC, Salisbury, UK), LoVo and BALB/3T3 were obtained from the American
Type Culture Collection (ATCC, Manassas VA, USA), and LoVo/DX cells were a gift from
Prof. Borowski from Gdańsk University of Technology, Poland. A549 and LoVo cells were
cultured in OptiMEM+RPMI (1:1) medium (PChO IIET PAS, Wroclaw, Poland)
supplemented with 5% fetal bovine serum (FBS) (GE Healthcare, Logan UT, USA), 2mM L-
glutamine (Sigma-Aldrich, Steinheim, Germany) and for LoVo cell lines also with 1mM
sodium puryvate (Sigma-Aldrich, Steinheim, Germany). Medium for LoVo/DX contained
additionally 10 µg/100 mL doxorubicin hydrochloride (Accord Healthcare Poland, Warsaw,
28

Poland). MCF-7 cells were grown in Eagle medium (PChO IIET PAS, Wroclaw, Poland)
supplemented with 10% FBS, 2mM L-glutamine, 1% amino acids, 8 µg/mL insulin (Sigma-
Aldrich, Steinheim, Germany). BALB/3T3 cells were cultured in DMEM (Gibco, Paisley,
UK) medium supplemented with 10% FBS (GE Healthcare, Logan UT, USA)), 2mM L-
Glutamine (Sigma-Aldrich, Steinheim, Germany). All culture media contained antibiotics:
100 U/mL penicillin and 100 µg/mL streptomycin (Polfa Tarchomin, Warsaw, Poland and
Sigma-Aldrich, Steinheim, Germany respectively). 24 hours before addition of a tested
compound, the cells were seeded in 96-well plates in respective culture media. Next, the
compounds were dissolved in DMSO (Avantor, Gliwice, Poland) and serially diluted in
culture medium in the range of 100 – 0.1 µg/mL and added to tested cells for 72 hours. The
proliferation inhibition was measured by means of sulforhodamine B (SRB) assay: cells were
fixed and permeabilized with 50 % cold trichloroacetic acid (Avantor, Gliwice, Poland) for 1
hour, next were washed with tap water, stained with 0.1 % SRB (Sigma-Aldrich, Steinheim,
Germany) in 1% acetic acid (Avantor, Gliwice, Poland) for 30 minutes, followed with
washing of unbound dye with the use of 1 % acetic acid. Following, SRB was extracted from
cells with 10 mM TRIS (tris(hydroxymethyl)aminomethane) (Sigma-Aldrich, Steinheim,
Germany) and the absorbance of resulting solution was measured at 540 nm wavelength in
universal plate reader Synergy H4 (BioTek Instruments, Winooski VT, USA). The results
were calculated as an IC50 – the concentration of tested agent, which inhibits proliferation of
50% of the cancer cells, using Cheburator 0.4, Dmitry Nevozhay software [Nevozhay D:
Cheburator Software for Automatically Calculating Drug Inhibitory Concentrations from In
Vitro Screening Assays. Plos One 2014, 9(9):10.]. Cells were also exposed to solvent DMSO
in the same serial dilutions as with the tested agents, and to cisplatin and doxorubicin
hydrochloride used as reference drugs (Accord Healthcare Poland, Warsaw, Poland).
Experiments were repeated at least three times in separate tests.
29